Inhalational Aesthetic Sevoflurane Exacerbates Eye Phenotype of SCA3 Transgenic Drosophila Model CURRENT

Background: Spinocerebellar ataxia type 3 (SCA3) is an autosomal dominant inherited neurodegenerative disease. The features of SCA3 include extremely short life expectancies, motor functions, and eye phenotypes. Sevoflurane is one of the most frequently used inhalational anesthetics and shows both neuroprotective and neurotoxic effects. Previous studies showed neurotoxicity of sevoflurane exposure to Alzheimer’s disease models. However, the effect of sevoflurane inhalation on SCA3 is not clear. Materials and Methods: Here, we exposed sevoflurane to SCA3-transgenic Drosophila model with clinically relevant concentrations and observed the consequent change of survival, motor function, and eye phenotype of the flies. Results: We found that sevoflurane exposure exacerbated eye phenotype but not survival or motor function of male SCA3-transgenic flies. The percentage of ommatidium retinal cell number of male SCA3-transgenic flies with 0%, 2.1%, or 3% of sevoflurane exposure was 70.2 ± 4.8%, 64.8 ± 4.5%, or 46.8 ± 2.9% respectively (ANOVA F = 27.86, total df = 10, p = 0.0002), while sevoflurane exposure did not show any harm to the control flies. Conclusions: Our results may acknowledge the need for caution of the potential hazard of sevoflurane application on patients with SCA3 or other poly-Q related neurodegenerative diseases.

to 86 repeats in one allele of affected patients and at-risk carriers. [2,3] Ataxin-3 may be an ubiquitin-binding protein that interacts with the valosin-containing protein and Rad23 for endoplasmic reticulum-associated degradation. [4][5][6] Expanded pathogenic polyQ proteins tend to aggregate with various target molecules, including proteasome subunits or various transcription factors, such as the TATA-binding protein (TBP) and the CREBbinding protein (CBP). [7,8] The abnormal CAG repeat of SCA3 is incurable, but treatments for relief are available for some symptoms. [9] The features of SCA3 include motor functions, such as gait problems and tremor, speech difficulties, clumsiness, as well as eye phenotypes such as frequent visual blurring and diplopia. [10] Besides, SCA3 patients have extremely short life expectancies. [11] Sevoflurane (2,2,2-trifluoro-1-[trifluoromethyl]ethyl fluoromethyl ether) is among the most frequently used inhalational anesthetics for general anesthesia during surgery. Anesthesia (2.1%) is a common clinically relevant concentration of sevoflurane, whereas 3.0% is a relatively high concentration. [12] Two opposite effects of sevoflurane have been reported, namely, the neuroprotective activity and neurotoxicity. On the one hand, Long-term exposure to sevoflurane could induce ER stress and further cause neuronal degeneration in aging rats [13] and could induce apoptosis and elevate Aβ levels, which may promote the neuropathogenesis of Alzheimer's disease. [14] Moreover, sevoflurane shows developmental neurotoxicity. [15][16][17] On the other hand, sevoflurane shows neuroprotective effects, such as the improvement of cognitive ability and the protection against damage to cerebral cortical neurons after brain injury, [18] and transient forebrain ischemia. [19] The reduction of the calcium-dependent glutamate release was thought to be the underlying mechanism of protection against neuronal injury by sevoflurane. [20] Since the inconclusive perspectives on the neurotoxic/neuroprotective role of sevoflurane, 4 its impact on neurodegenerative disease is not fully understood. It is clinically significant to decipher the effects of the most popular anesthetic agent. Although sevoflurane exposure increases apoptosis, inflammation, and Aβ levels in Alzheimer's disease (AD), [14,15,21] the effect of sevoflurane exposure on SCA3 is not clear.
In this study, we used Drosophila as our model to study whether sevoflurane ameliorates or exacerbate SCA3 phenotypes. Previous studies have shown that the pan-neuronal expression of human MJD1-Q84 with the pathogenic expanded polyQ stretch of 84 CAG repeats, but not the expression of human MJD1-Q27 with 27 CAG repeats, results in SCA3 phenotypes, such as shortened lifespan, attenuated motor function, and eye phenotype. [22,23] This was achieved by utilizing the GAL4-UAS system. [24] Specifically, the first filial generation of the offspring of the parental flies each carrying elav-Gal4 (panneuronal driver) and UAS-MJD1-Q84 (or UAS-MJD1-Q27 or w 1118 for control) was used as the SCA3 model and control respectively. Adult flies were used to avoid potential developmental neurotoxicity of sevoflurane exposure. Three doses of sevoflurane exposure, namely 0%, 2.1%, and 3%, were tested to simulate the clinical usage. Survival rate, motor function, and eye phenotype were selected as the indicators to judge the effectiveness of sevoflurane exposure on SCA3 Drosophila model.

Drosophila stocks and genetics
All stocks were obtained from the Bloomington Stock Center. Flies were raised on standard cornmeal food at 25 °C and 60% humidity in a 12 h light/dark cycle. The GAL4/UAS system was used for the overexpression of transgenic UAS in Drosophila as previously described [24]. Four strains of Drosophila were used: elav-Gal4, UAS-MJD1-Q84, UAS-MJD1-Q27, and w 1118 (wild type). Virgin female flies carrying the driver elav-Gal4 on the X chromosome were crossed with males carrying UAS-MJD1-Q84 or UAS-MJD1-Q27, or w 1118 .
All F1 offspring expressed Q27 or Q84 in the nervous system, thereby producing a model for SCA3. The virgin female flies carrying the driver elav-Gal4 were crosses with w1118 male flies and their F1 offspring were used as controls. The F1 offspring of elav-Gal4 > UAS-MJD1-Q84, abbreviated as elav > UAS-Q84 in the text, expresses an expanded polyQ stretch with 84 CAG repeats in the eye or nervous system inducing the pathogenic phenotypes in SCA3 patients, as previously described. [22] The F1 offspring of elav-Gal4 > UAS-MJD1-Q27, abbreviated as elav > UAS-Q27 in the text, expresses an expanded polyQ stretch with 27 CAG repeats, and does not present the disease phenotype. [25] The F1 offspring of elav-Gal4 > w 1118 , abbreviated as elav > w 1118 in the text, was used as another wild type control. Fly food was changed every 3-4 d to maintain the freshness of the surroundings.

Sevoflurane exposure
All flies were aged 6 days after eclosion and received 0%, 2.1% or 3% sevoflurane plus oxygen for 1 h in identical anesthetizing chambers. The flies were anesthetized once daily, and the process was performed for two times. Fly eye dissection and the tests of fly survival rate and climbing ability were performed on the day after the instances of anesthetization.

Fly eye dissection and immunohistochemistry
All flies were age-and sex-matched to assess the modification of eye phenotype. For immunocytochemistry, eyes samples from the flies anesthetized with temporary sevoflurane exposure (0%, 2.1% or 3.0%) for the subsequent 2 d were dissected and fixed in 4% paraformaldehyde in phosphate-buffered saline (PBS). Fixed samples were washed thrice in PBST (0.3% PBS with Triton X-100) for 10 min, and samples were incubated in the respective primary antibody in PBST with 5% goat serum at 4 °C overnight. Subsequently, the samples were washed thrice in PBST for 10 min and incubated in the secondary antibody in PBST with 5% goat serum at room temperature for 2 h. After incubation, samples were washed thrice for 10 min in PBST and mounted in 80% glycerol. Samples were analyzed under a Leica TCS SP2 confocal microscope. The procedures for wholemount adult fly retinal immunohistochemistry were described in previous studies. [26,27] The following primary antibodies were used in this study: lamin (1:20; DSHB) and

Statistical analysis
The number of phalloidin-stained photoreceptors was counted manually for statistical analysis. We calculated the total number of ommatidia then inferred the total number of 7 photoreceptors in the confocal sections by counting the remaining number of photoreceptors in the same confocal sections. The degenerative ratio is the ratio of the remaining photoreceptors to the inferred total number of photoreceptors (that is 7 X number of ommatidia). We used one-way ANOVA with Tukey's multiple comparison tests to determine the significant values. Statistical significance was set at p < 0.05. For the comparison of survival or climbing curves, the log-rank (Mantel-Cox) test was used.

Exposure to sevoflurane did not influence survival rate in SCA3-transgenic flies
To access whether sevoflurane ameliorates or exacerbate SCA3 phenotypes, the SCA3 Drosophila model of elav>Q84 and two control groups, namely elav>Q27 and elav>w1118, were treated with 0%, 2.1%, and 3% of sevoflurane and observed for subsequent survival rate, motor function, and eye phenotype for both male and female respectively. For survival rate, as an internal control of 0% sevoflurane treatment, the SCA3 model showed significantly deteriorated survival for both male (Fig. 1A, Supp. Table 1) and female (Fig.   1B, Supp. Table 2) compared to both control groups. However, both doses of sevoflurane neither ameliorated nor exacerbated survival of SCA3 model for both male (Fig. 1C, Supp. Table 1) and female (Fig. 1D, Supp. Table 2) compared to 0% control groups. Sevoflurane treatment also showed no harm to control groups with these dosages for both male (Supp. Fig. 1A and 1B) and female (Supp. Fig. 1C and 1D).

Exposure of sevoflurane did not influence the motor function of SCA3-transgenic flies
For motor function, as an internal control of 0% sevoflurane treatment, the SCA3 model showed significantly deteriorated climbing ability for both male ( Fig. 2A, Supp. Table 3) and female (Fig. 2B, Supp. Table 4) compared to both control groups. As the case of survival rate, both doses of sevoflurane neither ameliorated nor exacerbated climbing 8 ability of SCA3 model for both male (Fig. 2C, Supp. Table 3) and female (Fig. 2D, Supp. Table 4) compared to 0% control groups. Sevoflurane treatment also showed no harm to control groups with these dosages for both male (Supp. Fig. 2A and 2B) and female (Supp. Fig. 2C and 2D).

Exposure of sevoflurane exacerbated eye phenotype of male SCA3-transgenic flies
For eye phenotype, the completeness of ommatidium retinal cell was quantified by counting the cell number of ommatidium retinal cell normalized to 7 times of ommatidium number, since seven retinal cells can be observed in a wild type ommatidium on a confocal section. Thus, a 100% ratio denotes a normal phenotype, while a 50% ratio indicates an ommatidium with only half retinal cells averagely. As an internal control of 0% sevoflurane treatment, the SCA3 model showed significantly deteriorated eye phenotype with nearly 30% or 40% for male (Fig. 3A left panel) or female (Fig. 3C left panel) respectively compared to both control groups. Moreover, sevoflurane treatment showed no harm to control groups with these dosages for both male ( Fig. 3A the first two rows, and 3B) and female ( Fig. 3C the first two rows, and 3D). Interestingly, sevoflurane treatment significantly exacerbated eye phenotype in the male SCA3 Drosophila model in a dose-dependent manner ( Fig. 3A bottom row, 3B, and Table 1), but not the female counterpart ( Fig. 3C bottom row, 3D, and Supp. Table 5).

Discussion
In this study, we found that sevoflurane exposure with clinically relevant concentration exacerbated eye phenotype but not survival or motor function of SCA3-transgenic flies.
Furthermore, this phenomenon is sex-dependent and restricted to male flies. Our findings identified heterogeneous effects of sevoflurane exposure on distinct SCA3 phenotypes, and this may imply a board set of indicators, but not a single one should be used to judge 9 the impact of a treatment on a disease. Since flies are much more tolerant to ischemic anesthesia, [28,29] our results may acknowledge the need for caution of the potential hazard of sevoflurane application on patients with SCA3 or other poly-Q related neurodegenerative diseases. Meanwhile, for the wild type or overexpression of normal MJD1 models, sevoflurane exposure showed no harm to survival, motor function, and eye phenotype. These results confirmed the safety of sevoflurane application on healthy subjects.
The sex-biased effect of sevoflurane exposure on SCA3 flies may be caused by the bias of the expression of MJD1-Q84, rather than the difference in the hormone system. A previous study of SCA3 using a similar Drosophila model of SCA3 reported much more severe phenotype in the male SCA3 model and identified the X chromosome dosage compensation to be the cause of this consequence. [25] Although the genetic instability of CAG repeat units in male SCA3 patients has been mentioned, [30] this phenomenon might be not universal. [31] Therefore, this may imply that the potential hazard of sevoflurane on patients with SCA3 is not sex-dependent, but MJD1-polyQ expression dependent.
In contrast to our previous findings of protective effects of sevoflurane on AD Drosophila model, [32] this study identified the deleterious one on SCA3 flies. These opposite effects highlight the complex nature of sevoflurane on neurodegenerative diseases. Further investigations are essential to deepen our understanding in this field to ultimately solve the long-standing paradoxical role of sevoflurane on neurotoxicity.
Previous studies identified sevoflurane exposure decreases extracellular signal-regulated kinase (ERK) phosphorylation; however, this event induces toxicity in the developing but not adult brain. [33,34] Follow-up studies showed that sevoflurane-induced cognitive dysfunction could be rescued by regulation of Tau/GSK3β and ERK/PPARγ/CREB signaling, [35] and PPARγ dysregulation may induce microglia-mediated neuroinflammation. [36] Another line of evidence showed that reduction of the glial cell-derived neurotrophic factor might be the cause of the anesthesia-induced cognition deficits. [37] These shreds of evidence address the ERK-PPARγ-microglia cascade to the neurotoxicity induced by sevoflurane exposure. However, preconditioning with sevoflurane or isoflurane showed protective effects against ischemic stress, [38,39] probably through inducible NO synthesis [40] and activation of mitochondrial ATP-sensitive potassium channels. [41] In this study, we showed that sevoflurane exposure exacerbated SCA3 phenotype; however, the relevance of the ERK-PPARγ-microglia cascade has not been addressed in SAC3 pathology and is a target for future research.

Conclusions
In conclusion, we identified that sevoflurane exposure with clinically relevant

Ethics approval and consent to participate
Not applicable.

Consent for publication
Not applicable.

11
The data generated and/or analyzed during the current study are available from the corresponding author on reasonable request

Competing interests
The authors certify that there is no competing interest with any financial organization regarding the material discussed in the manuscript.

Funding
This work was supported by grants from the Ministry of Science and Technology in Taiwan

Authors' contributions
C-W C. and W-Y L. designed the model and the computational framework and analyzed the data. K-B C. and Y-C K. carried out the implementation. J.C. performed the calculations and wrote the manuscript with input from all authors. C-Y L. and H-P L. conceived the study and were in charge of overall direction and planning.  Exposure to 2.1% or 3.0% sevoflurane exacerbated SCA3 eye phenotype in male flies. (A-B) Exposure to 2.1% or 3.0% sevoflurane did not cause any eye phenotype in control flies (elav>w1118, and elav>Q27), but exacerbated SCA3